Research On Load Optimal Allocation And Decision Making In Thermal Power Plants

With the deepening of power industry reform in China and the implementation of "separation of plant and network, electricity bidding" mechanism, the power generation enterprises has changed over from the production type enterprises to operation enterprises, the power plant participates in market competition as an independent economic entity. The power network no longer distributes load instruction on power unit, but to all units of a power plant under the total load instruction. Therefore, the scientific and rational allocation for each unit of the load in the plant will be conducive to optimizing the unit operation level, reduce the coal consumption of power plant, and provide scientific basis for power plants in power market bidding. At the same time, energy saving and emission reduction of enterprises and the sustainable development policies of the state, also brought new requirements to the power enterprises. Thermal power plants produce large amounts of harmful gases in the power generation process, which cause serious pollution to the environment. Pollutant emissions have become non-negligible factors of load distribution in thermal power plant. Therefore, it is significant to study on the load optimal allocation of thermal power plants.This paper introduces the basic concept of load optimal allocation in thermal power plants firstly, then according to domestic and foreign research status of thermal power plant's load optimal allocation, combined with the actual operation of the power plant, this paper chooses net coal consumption as the economic evaluation index of load allocation and determines that the coal consumption characteristic curve of fossil-fired unit is the basis of optimal load dispatch of power plant. The coal consumption characteristic curve of the unit is fitted by least square method. A multi-objective mathematical model based on environment and economy is established.In the study of single objective load optimal allocation algorithm, this paper establishes a single-objective mathematical model of load optimal allocation based on economy by using the coal consumption characteristic curves of fossil-fired units. Then this paper describes Equal Incremental method and dynamic programming in details and compares their characteristics, using dynamic programming method as single objective optimization algorithm, and gives a detailed description and design for the algorithm. The experimental results testify that dynamic programming method can optimize the operation of unit to a certain extent, thereby reducing the coal consumption, and ensuring the economy of thermal power plant's load allocation.Based on single-objective mathematical model and consider the emission reduction targets, this paper establishes the multi-objective mathematical model of load optimal allocation based on environment and economy and designs an algorithm based on adaptive grid particle swarm optimization algorithm (AGC-MOPSO) to find the Pareto-optimal solutions and uses Multiple Attribute Decision Making (MADM) to give the most satisfactory results; The design of AGC-MOPSO includes:treatments for equality and inequality constraints; The maintenance of Pareto external file based on adaptive grid method; The selection of personal best position and global best position. This paper uses the calculation example and compares the experimental results with the algorithm based on genetic algorithms to demonstrate the validity of the method designed in the paper. For the Multiple Attribute Decision Making (MADM) of the Pareto-optimal solutions, this paper use information entropy which is an objective weighting method to calculate the weight of environment and economy, and then ranks the Pareto-optimal solutions by ideal solution method (TOPSIS) to get the most satisfactory solution.This paper also applies the above methods to the production operation management system of a thermal power plant. This paper builds the system framework based on Flex+Spring+Hibernate, designs a simple and beautiful interface, realize the unit load data base management, coal consumption characteristic curve fitting, set of single-objective load optimal allocation and multi-objective load optimal allocation. It improves the enterprise's management and production efficiency, and brings economic benefit for the enterprise.